Motif Design for Nitrilases Nikhil Sharma and Tek Chand Bhalla* Sub- Distributed Information Centre, Himachal Pradesh University, Summer-Hill, Shimla-171005, India

g in Geno nin m i ic M s

ta & a P Sharma and Bhalla, J Data Mining Genomics Proteomics 2012, 3:3 D r

f o Journal of

o t

i DOI: 10.4172/2153-0602.1000119

c u

s o J ISSN: 2153-0602 Data Mining in Genomics & Proteomics

Research Article Open Access Motif Design for Nitrilases Nikhil Sharma and Tek Chand Bhalla* Sub- Distributed Information Centre, Himachal Pradesh University, Summer-Hill, Shimla-171005, India

Abstract Nitrilase is one of the nitrile metabolizing enzymes that catalyses the conversion of nitriles to corresponding acids which has gained importance in green chemistry. Nitrilase being substrate specific yet it acts over a wide range of nitriles (aliphatic/aromatic) has drawn attention due to its utility in mild hydrolysis. Most of the nitrilases reported hitherto have been physically extracted characterized from the microbial/plant sources. In order to identify sequences for nitrilases two groups of motif were designed i.e. aliphatic nitrilase motif’s (MDMAl) and aromatic nitrilase motif’s (MDMAr) each with four motifs specifically belonging to nitrilase with conserved catalytic triad (Glu-48, Lys-131, Cys-165) which can be used as marker for nitrilase. Conserved regions were identified by performing Multiple Sequence Alignment (MSA) using Multiple EM for Motif Elicitation (MEME). The Manually Designed Motifs (MDM’s) were validated by ScanProsite and their presence is also confirmed by PRATT, Gblocks and MEME. The ScanProsite search against the MDMAr exhibited some new sources of aromatic nitrilase from plant, animals and microbes whereas MDMAl only exhibited nitrilase from microbes. Besides identifying unique motifs in order to confirm their substrate specificity for nitriles, randomly selected sequences were validated by studying some important physiochemical parameters and position specific amino acids.

Keywords: Gblock; Manually designed motif (MDM); Multiple of the aromatic and aliphatic nitrilases [13] and using bioinformatics alignments; Nitrilase; PROSITE; Pratt; Multiple EM for motif elicitation approaches, motifs were manually designed to differentiate and (MEME); ScanProsite; Substrate specificity identified aromatic and aliphatic nitrilases. Computational analysis of amino acid sequences and study pertaining to physiochemical Background properties [14] of nitrilases have revealed differences between aromatic Nitrilases are responsible to catalyze the hydrolysis of nitriles into and aliphatic nitrilases [13]. corresponding acids and ammonia which are used in chemistry for the production of industrially important acids. Nitrilases frequently exhibit Methodology enantioselectivity under mild reaction conditions, making them ideal Retrieval of sequences and designing of motifs tools for green chemistry from conversion of nitrile or in environment management for remediation of nitrile contaminated soil, water and The protein sequences were obtained from the protein server air. Based on substrate specificity they are grouped as aromatic and UniprotKB/SwissProt (release 2011-12) database. To design motifs aliphatic nitrilase. manually, six microbial aliphatic and aromatic nitrilases were selected on the basis of our earlier study (Table 1) [13]. A large number of microbial plant and animal genome have been sequenced in the past decade and genome sequence data have Multiple Sequence alignment and phylogenetic analyses tremendously expanded over those years. Screening of genome/ Multiple sequence alignments were performed using CLUSTAL W proteome databank will be worthwhile to find out novel sources of [15] and CLUSTAL X (version 2.1) software with default settings and enzymes. Many tools and techniques such as BLAST, Hidden Markov verified with MAFFT (Multiple Alignment with Fast Fourier Transform) Model (HMM), neural network classification are used for in silico and MEME (Multiple Em for Motif Elicidation). Nitrilase (aliphatic screening of genome/proteome sequence database till date. Among and aromatic) sequences were used for phylogenetic relationship these, motif design has been found to be one of the most reliable inferences. Phylogenetic tree was generated by Neighbor Joining (NJ) strategy for efficient screening of database [1-3] as motifs of a particular using CLUSTAL X (Figure 2). After multiple alignments and also on protein sequence signifies its specific structure and functionally which the basis of presence of catalytic triad i.e glutamine, lysine, and cysteine is useful to characterize and classify that protein. (Glu-48, Lys-131, and Cys-165) motifs were manually designed (MDM) Nitrile metabolizing microorganisms are mostly isolated from soil for aliphatic (MDMAl) and aromatic nitrilases (MDMAr). In order to or water using enrichment culture method and these are cultured and verify the manually designed motif and to eliminate poorly aligned or tested for the nitrilase activity during screening of microbial isolates. This is a classical method for isolation of the nitrile metabolizing microorganisms and subsequent screening for nitrilases activity is a time *Corresponding author: Dr. Tek Chand Bhalla, Sub-Distributed Information consuming and cost intensive process [4-7]. Presently bioinformatical Centre, Himachal Pradesh University, Summer-Hill, Shimla-171005, India, E-mail: tools and techniques such as Blast [1], HMM (Hidden Markov Model) [email protected] [8-9] the neural network classification [10,11] and Motif Identification Received August 06, 2012; Accepted September 24, 2012; Published October Neural Design (MOTIFIND) [12] are used for screening of genome 01, 2012 and sequence database for searching gene coding for novel enzymes. Citation: Sharma N, Bhalla TC (2012) Motif Design for Nitrilases. J Data Mining Genomics Proteomics 3:119. doi:10.4172/2153-0602.1000119 The present communication aims at Manually Designed Motif (MDM) for aliphatic and aromatic nitrilase and their validation using Prosite, Copyright: © 2012 Sharma N, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted ScanProsite, BLAST, PRATT and G-block is being reported. On the use, distribution, and reproduction in any medium, provided the original author and basis of the earlier studies on in silico analysis of amino acid sequence source are credited.

J Data Mining Genomics Proteomics ISSN:2153-0602 JDMGP an open access journal Volume 3 • Issue 3 • 1000119 Citation: Sharma N, Bhalla TC (2012) Motif Design for Nitrilases. J Data Mining Genomics Proteomics 3:119. doi:10.4172/2153-0602.1000119

Page 2 of 6

Microorganism Substrate specificity Accession Number In-silico physiochemical characterization and validation of (ExPASy) motifs R. rhodochrous J1 Aliphatic Q03217 R. rhodochrous K22 Aliphatic Q02068 After confirmation through MEME version 4.8.1 (Figure 1a Methylibium petroleiphilum Aliphatic A2SEG6 and Figure 1b), Gblock tool, Manually Designed Motif’s (MDM’s) Pseudomonas syringae pv. syringae Aliphatic Q500U1 Synechococcus sp. ATCC 27144 Aliphatic Q5N478 were subjected to database search through ScanProsite to validate Bradyrhizobium sp ORS278 Aliphatic A4YWK0 and determine the specificity of these motifs in identifying nitrilase Burkholderia cepacia J2315 Aromatic B4EE44 sequences. ScanProsite tool was used to search the databases both from Pseudomonas entomophila Aromatic Q1I7X1 Janthinobacterium Marseille Aromatic A6T0X3 UniprotKB/Swiss-Prot and UniprotKB/TrEMBL with match mode Microscilla marina Aromatic A1ZD79 of “not greedy, not overlaps” and “no includes”. ScanProsite analysis Bordetella bronchiseptica Aromatic Q7WNC4 resulted in protein sequences which were counted and were grouped Shewanella sediminis Aromatic A8FQL4 into plants, bacteria and uncultured organisms (Table 2). Table 1: Names of various microrganisms and accession number of the some microbial nitrilases (aliphatic and aromatic). MDM were further verified by using Pratt tool to identify

Figure 1a: Aliphatic Motif’s for Nitrilase validated by MEME version 4.8.1(2012) (the height of the motif “block” is proportional to -log(p-value), truncated at the height for a motif with a p-value of 1e-10). Figure 1b: Aromatic Motif’s for Nitrilase validated by MEME version divergent regions of aligned protein Gblock software was used [16,17]. 4.8.1(2012) (the height of the motif “block” is proportional to -log(p-value), truncated at the height for a motif with a p-value of 1e-10). Bootstrap value was calculated by using default value (1000).

Page 3 of 6

Nitrilases Manually Designed Motif Protein Sequence Obtained Hits obtained for MDM Motif Presence UniProtKB/TrEMBL Entries [FL]-[ILV]-[AV]-F-P-E-[VT]-[FW]-[IL]-P-[GY]-Y-P-[WY] F-17 84 B-38 34-68 84 UO-29 Aliphatic R-R-K-[LI]-[KRI]-[PA]-T-[HY]-[VAH]-E-R F-31 115 B-53 125-177 115 UO-31 C-W-E-H-[FLX]-[NQ]-[PT]-L F-49 248 B-132 157-215 248 UO-67 [VA]-A-X-[AV]-Q-[AI]-X-P-[VA]-X-[LF]-[SD] F-06 130 B-87 1-30 130 UO-37 [ALV]-[LV]-[FLM]-P-E-[AS]-[FLV]-[LV]-[AGP]-[AG]-Y-P F-01 B-42 55 26-55 55 P-10 UO-2 [AGN]-[KR]-H-R-K-L-[MK]-P-T-[AGN]-X-E-R F-09 B-68 104 P-18 165-206 104 UO-8 Aromatic A-01 C-W-E-N-[HY]-M-P-[LM]-[AL]-R-X-X-[ML]-Y F-3 B-67 128 125-180 128 A-01 UO-23 A-X-E-G-R-C-[FW]-V-[LIV] F-19 B-74 105 191-216 105 A-05 UO-08 Table 2: Manually designed motif’s with number of hits obtained from UniProt KB and TrEMBL and their position. F-Fungus, B-Bacteria, UO-Uncultured Organism, P-Plants and A-Animal.

Pattern 1 ([FL]-[ILV]-[AV]-F-P-E-[VT]-[FW]-[IL]-P-[GY]-Y-P-[WY]) S. No Organism Access. No. Mol wt. Amino Acid No. NCR* Alanine (%) Instability index 1 Uncultured organism Q6RWK6 378791 354 39 16.9 36.02 2 Nocardia sp. Q5D4V8 41971.9 381 51 11.5 30.81 3 Sclerotinia sclerotiorum A7F824 41160.5 368 49 6.8 35.31 4 Nectria haematococa C7YVF8 41018.6 366 53 7.7 33.84 5 Silicibacter pomeroyi Q5LLB2 37226.1 344 44 10.8 35.63 Pattern 2 (R-R-K-[LI]-[KRI]-[PA]-T-[HY]-[VAH]-E-R) 1 Uncultured organism Q6RWS5 37242.6 345 40 15.4 37.02 2 Uncultured organism Q6RWS0 38048.3 349 36 12.3 35.09 3 Rhodococcus rhodochorus A4LA85 40370.4 366 46 10.9 39.90 4 B. japonicum Q89GE3 36203.3 334 39 13.5 33.69 5 Aspergillus niger A9QXEO 40022.2 356 49 7.6 30.71 Pattern 3 (C-W-E-H-[FLX]-[NQ]-[PT]-L) 1 Burkholderia multivorans B9BNU9 37461.8 345 42 11.9 36.48 2 Uncultured organism Q6RWF2 37809.2 344 44 11.0 39.69 3 Uncultured organism Q6RWF9 38029.4 353 40 17.0 37.41 4 Methylibium petroleiphilum A2SEG6 37969.1 357 39 16.0 42.17 5 Paracoccidioides brasiliensis C1H5H4 37871.1 356 39 10.1 37.47 Pattern 4 ([VA]-A-X-[AV]-Q-[AI]-X-P-[VA]-X-[LF]-[SD]) 1 Rhodococcus rhodochrous Q03217 40189.2 366 46 11.2 39.36 2 Uncultured organism Q6RWK6 37879.1 354 39 16.9 36.02 3 Sphingomonas wittichi B4WRX6 35995.2 334 42 12.6 38.51 4 Silicibacter pomeroyi Q5LLB2 37226.1 344 44 10.8 35.63 5 Providencia alcalifaciens B6XD88 37552.7 343 43 10.8 30.32

Table 3a: In silico analysis of some physiochemical properties of some aliphatic nitrilase (sequences obtained from ScanProsite) from Manually Designed Motif analysis. conserved pattern in protein sequences. In order to find the nature of residue (Asp+Glu), molecular weight, alanine content (%) and the nitrilase sequences i.e. aliphatic/aromatic we also studied some of instability index. The values of these parameters for the predicted/ the physiochemical properties of randomly selected protein sequences. reported nitrilase sequences were deduced from the ProtParam The physicochemical parameters studied included negatively charged (http://web.expasy.org/protparam/) of Expert Protein Analysis System

Page 4 of 6

Pattern 1 ([ALV]-[LV]-[FLM]-P-E-[AS]-[FLV]-[LV]-[AGP]-[AG]-Y-P) S. No Organism Access. No. Mol wt. Amino Acid No. NCR* Alanine (%) Instability index 1 Burkholderia capacia HI2424 A0K4N0 32737.0 307 36 11.4 41.50 2 Microscilla marina A1ZD79 33583.0 302 31 7.9 42.07 3 Magnoparthe grisea A4R7D5 34847.0 324 42 9.0 44.68 4 Populus trichocarpa B9HBW3 28464.0 266 28 10.5 41.42 5 Pseudomonas aeruginosa A6V5Q2 33417.2 310 35 10.6 40.34 Pattern 2 ([AGN]-[KR]-H-R-K-L-[MK]-P-T-[AGN]-X-E-R) 1 Shewanella sediminis (strain HAW-EB3) A8FQL4 34872.8 317 41 8.5 41.24 2 Actinobacillus minor NM305 C5RYV4 33992.0 307 39 7.5 43.43 3 Pirellula staleyi DSM 6068 D2R9H8 32521.3 302 35 10.6 49.18 4 Pantoea sp. At-9b. C8Q5J0 33273.0 306 35 10.5 40.97 5 Burkholderia cenocepacia (strain AU 1054) Q1BZ21 32737.4 307 36 11.4 41.50 Pattern 3 (C-W-E-N-[HY]-M-P-[LM]-[AL]-R-X-X-[ML]Y) 1 Aspergillus niger A2QYH5 34964.7 320 36 8.8 41.75 2 Shewanella pealeana A8H6J4 34786.8 314 39 9.2 39.82 3 Syntrophobacter fumaroxidans A0LKP2 36011.4 328 44 11.0 41.02 4 Burkholderia ambifaria B1FFB0 32904.5 307 38 11.1 40.05 5 Algoriphagus sp. PR A3HXT3 34738.1 305 43 4.9 41.30 Pattern 4 (A-X-E-G-R-C-[FW]-V-[LIV]) 1 Syntrophobacter fumaroxidans A0LKP2 36011.4 328 44 11.0 41.02 2 Talaromyces stipitatus B8MN39 35560.5 325 41 10.2 46.51 3 Uncultured organism Q6RWR5 33711.5 310 36 11.0 41.86 4 Microscilla marina A1ZHQ2 34937.1 311 38 9.0 41.02 5 Rhodopseudomonas palustris D2MB94 33698.4 317 40 13.9 39.93

Table 3b: In silico analysis of some physiochemical properties of some aromatic nitrilase (sequences obtained from ScanProsite) from Manually Designed Motif analysis.

(ExPASy) i.e. the proteomic server of Swiss Institute of Bioinformatics 160-180 and 1-30 amino acid numbers respectively (see additional file (SIB). Fasta format of sequences were used for analysis. 1). Presence of MDMAl through Pratt analysis showed that the first, second and third DMP (design motif pattern) were in between 34 to Results 68, 125 to177 and 157 to 215 numbered amino acids, respectively (see Manually Designed Motifs were validated for the two groups of additional file 1). nitrilases i.e. aliphatic and aromatic and the results are shown in Table 1. Total amino acid numbers of all resulted sequences were found to Hits obtained from each manually designed motif (MDM) of aliphatic be in between 330-382. Molecular masses were from 37226 to 41971 and aromatic nitrilases in SwissProt and TrEMBL databases are obtained daltons, alanine content was in between 10-16%, negative charged through a motif search. These motifs are conserved for aliphatic and residue were in between 37-52 and instability index ranged from 30.32 aromatic nitrilases (Figure 1a and Figure 1b). MEME version 4.8.1 to 44.13 for the aliphatic nitrilase sequences (Table 3a). (2012) software also confirmed the presence of all the MDM, which were analyzed through ScanProsite and resulted in protein sequences Designing and validation of motifs for aromatic nitrilase only for nitrilase family. These sequences have conserved catalytic triad Manually designed motifs for aromatic nitrilase (MDMAr) were i.e. glutamine, lysine, and cysteine (Glu-48, Lys-131, Cys-165) which (i) [ALV]-[LV]-[FLM]-P-E-[AS]-[FLV]-[LV]-[AGP]-[AG]-Y-P; (ii) are essential for the nitrilases activity [18-21]. [AGN]-[KR]-H-R-K-L-[MK]-P-T-[AGN]-X-E-R; (iii) C-W-E-N- Designing and validation of motifs for aliphatic nitrilase [HY]-M-P-[LM]-[AL]-R-X-X-[ML]Y and (iv) A-X-E-G-R-C-[FW]-V- [LIV]. ScanProsite analysis of all the four MDMAr resulted in 55, 104, In the present study manually designed motifs for aliphatic ni- 128 and 105 protein sequences respectively. Out of these 392 protein trilases (MDMAl) were (i)[FL]-[ILV]-[AV]-F-P-E-[VT]-[FW]-[IL]-P- sequences, 251 were of microbes, 28 of higher plants and 7 from [GY]-Y-P-[WY]; (ii)R-R-K-[LI]-[KRI]-[PA]-T-[HY]-[VAH]-E-R; (iii) animals. Forty one sequences belonged to uncultured organisms. Plant C-W-E-H-[FLY]-[NQ]-[PT]-L and (iv) [VA]-A-X-[AV]-Q-[AI]-X- protein sequences were only obtained from first and second MDMAr P-[VA]-X-[LF]-[SD]. ScanProsite analysis of these MDMAl resulted and not from (iii) and (iv) manually designed motifs (Table 3). The in 85 (i) 115 (ii) 248 (iii) 130 and (iv) hits of amino acid sequences designed motif pattern was found to be in between 40-60; 130-165; 160- respectively. All the 578 protein sequences those were obtained after 185; 200-220 amino acid respectively (see additional file 2). In order to ScanProsite analysis of all the four MDMAl pertained to microbes only. confirm the ScanProsite results, Pratt analysis regarding the presence In these 578 protein sequences, there were 164 nitrilase sequences from of MDM was done and it was found that first, second, third and fourth the uncultured microorganisms as shown in Table 2. designed motif pattern (DMP) were in between 26 to 55, 125 to 180, 165 to 206 and 191 to 216 amino acids, respectively. MDMAl were found between amino acid (i) 40-55, (ii) 125-140, (iii) 160-180, (iv) 1-30. All these results except the fourth MDMAl were In silico physiochemical analysis of the protein sequences retrieved further confirmed by the Pratt analysis. According to Pratt analysis first, from ScanProsite using first, second, third and fourth MDMAr second, third and fourth MDMAl were found between 40-55, 125-140, exhibited the total number of amino acid ranged between 306-331,

Page 5 of 6

and reliable nitrilase identification and compared it to the currently available methods, including the BLAST search, the PROSITE pattern search, Gblock [23] MEME version 4.8.1 (Multiple Em for Motif Elicitation) and the HMM method. All the designed motifs have one amino acid residue of active site, therefore when MDM were analyzed and validated through ScanProsite, resulting protein sequences were specifically belonging to nitrilase with conserved catalytic triad Glu-48, Aromatic Lys-131, Cys-165 [18,24,20,15] and total number of hits were found to Nitrilases be significantly higher (578 aliphatic & 392 aromatic) when compared to databases such as BLAST and PROSITE. Gblock eliminated poorly aligned and divergent region of the protein sequences for better phylogenetic analysis. Pratt results confirmed that the Manually Designed Motif for aliphatic nitrilase (MDMAl ) ranged between 40-55 amino acid for pattern 2; 125- 141 amino acid for pattern 3; 160-172 amino acid for pattern 4 (see additional file 1 & 2). This revealed that Pratt analysis for MDM are reliable and accurate tool for conducting search of similar sequence from databases as its results were similar to the Scan Prosite analysis as other similar sequence from databases as its results were similar to the ScanProsite analysis. One motif i.e. [VA]-A-X-[AV]-Q-[AI]-X-P-[VA]- X-[LF]-[SD] was not confirmed by the Pratt analysis as Pratt started the th Aliphatic analysis after 30 amino acid and we have designed this MDM between

Nitrilases 1-30 amino-acid (Table 2). Multiple sequence alignment (MSA) in present study revealed that there are specific amino acids present at specific position which are responsible for the activity of nitrilases. Phylogenetic analysis by the Neighbor Joining (NJ) method clearly makes a distinction between the two groups of nitrilases i.e. aliphatic and aromatic nitrilases (Figure 2). Figure 2: Phylogenetic tree of some nitrilase protein sequences from In our previous communication [13] we have reported that some aliphatic and aromatic bacterial source organism constructed by NJ method, where aliphatic and aromatic is shown in two different groups. of the physiochemical properties play a significant role in the substrate specificity of aliphatic and aromatic nitrilases. The physicochemical parameters analysis of these sequences further confirmed that all the molecular sizes were found to be between 32273-36940 dalton’s, alanine sequences having a total number of amino acid between 330-382, content was between 4.9-11.0%, negative charged residues ranged from molecular weight between 37226–41971 daltons, alanine content 10- 34-44 and instability index was found to be 39.82-48.68 for the aromatic 16%, negatively charged residue (Asp+Glu) 37-52 and instability index nitrilase sequences (Table 3b). 30.32-44.13 belonged to aliphatic nitrilases. Similarly sequences having Phylogenetic analysis a total number of amino acid between 300-330, molecular weight between 32273–36940 daltons, alanine content (%) 8-11%, negative Twelve protein sequences for nitrilases (aliphatic/aromatic) charged residue (Asp+Glu) 37-52 and instability index 39.82-48.68 were subjected for the phylogenetic tree construction. Tree topology were of aromatic nitrilases (Table 3a and 3b). generated by Neighbor Joining Method (NJ) revealed that aliphatic and aromatic nitrilases were clustered with well distinct groups and well It is well known that uncultured organisms are the new sources supported bootstrap value of 1000 (Figure 2). of enzymes, antibiotics and drug discovery [25,26] In the present study, we have found nitrilase sequences belonging to uncultured Discussion organisms from the database (aliphatic-164, aromatic-41). The protein The expansion of molecular sequences and genomic databases has sequences of these uncultured organisms have all the specific properties made the area of genome sequence data analysis more challenging and responsible for the nitrilase activity. Further studies on these sequences interesting to develop tools for rapid and reliable searching and analysis may lead to find out specific nitrilases needed for transformation of of the data. In this endeavor, database searching against gene/protein nitriles in organic chemistry and industry. families with motifs has emerged as important strategy for efficient similarity searching [22]. While several domain/motif databases are Conclusion being compiled, it is important to develop database search tools that Nitrilases are distributed among microorganisms, plants and some fully utilize the conserved structural and functional information plant species. We here present computational analysis of two classes of embedded in those sequence data to enhance the reliability of the nitrilase enzyme which includes the study of motifs, physiochemical search. Sequence motifs typically occur in a specific and known order and phylogenetic analysis which could be used as tool to predict and in a sequence family. The ordering and spacing of motifs therefore, differentiate nitrilases on basis of their substrate specificity. The present provide powerful additional criteria for classifying sequences into analysis predicts new sources of nitrilases with study of common and families. In this paper, we have manually designed groups of motifs important physicochemical/biochemical properties as they share a (MDM) i.e. MDMAl and MDMAr each with four motifs for rapid common ancestry and use of these observations will be useful to predict

Page 6 of 6 molecular function of the randomly selected nitrilases. Computational 12. Wu CH, Zhao S, Chen HL, Lo CJ, McLarty J (1996) Motif identification neural analysis of various properties of nitrilases revealed differences and also design for rapid and sensitive protein family search. Comput Appl Biosci 12: 109-118. MEME, Pratt and Gblock analysis have confirmed the presence of four motifs. Additionally this approach has also led us to find new sources 13. Sharma N, Kushwaha R, Sodhi JS, Bhalla TC (2009) In silico analysis of amino of nitrilase across SwissProt/TrEMBL databases. MDM for aliphatic/ acid sequences in relation to specificity and physiochemical properties of some microbial nitrilases. J Proteomics Bioinform 2: 185-192. aromatic nitrilases, have been validated with the results and these motifs will be of potential application for screening genome/protein 14. Rabilloud T, Adessi C, Giraudel A, Lunardi J (1997) Improvement of the sequence databases to find novel sources of nitrilases. solubilization of proteins in two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 18: 307-316.

Acknowledgement 15. Thompson JD, Plewniak F, Poch O (1999) BAliBASE: a benchmark alignment The authors are grateful to Department of Biotechnology, Ministry of Science database for the evaluation of multiple alignment programs. Bioinformatics 15: and Technology, Govt. of India for funding. 87-88. 16. Castresana J (2000) Selection of conserved blocks from multiple alignments for References their use in phylogenetic analysis. Mol Biol Evol 17: 540-542.

1. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local 17. Talavera G, Castresana J (2007) Improvement of phylogenies after removing alignment search tool. J Mol Biol 215: 403-410. divergent and ambiguously aligned blocks from protein sequence alignments. Syst Biol 56: 564-577. 2. Henikoff S,Henikoff JG (1991) Automated assembly of protein blocks for database searching. Nucleic Acids Res 19: 6565-6572. 18. Pace HC, Brenner C (2001) The nitrilase superfamily: classification, structure 3. Attwood TK, Beck ME, Bleasby AJ, Parry-Smith DJ (1994) PRINTS--a database and function. Genome Biol 2: S001. of protein motif fingerprints. Nucleic Acids Res 22: 3590-3596. 19. Brenner C (2002) Catalysis in the nitrilase superfamily. Curr Opin Struct Biol 4. Steele HL, Jaeger KE, Daniel R, Streit WR (2009) Advances in recovery of 12: 775-782. novel biocatalysts from metagenomes. J Mol Microbiol Biotechnol 16: 25-37. 20. O’Reilly C, Turner PD (2003) The nitrilase family of CN hydrolyzing enzymes-a 5. Lefevre F, Jarrin C, Ginolhac A, Auriol D, Nalin R (2007) Environmental comparative study. J Appl Microbiol 95: 1161-1174. metagenomics: An innovative resource for industrial biocatalysis. Informa Healthcare 25: 242-250. 21. Chen CY, Chiu WC, Liu JS, Hsu WH, Wang WC (2003) Structural basis for catalysis and substrate specificity of Agrobacterium radiobacter N-carbamoyl- 6. Cowan DA, Arslanoglu, A, Burton SG, Baker GC, Cameron RA, et al. (2004) D-amino acid and amidohydrolase. J Biol Chem 278: 26194–26201. Metagenomics, gene discovery and the ideal biocatalyst. Biochem Soc Trans. 32: 298-302. 22. Altschul SF, Boguski MS, Gish W, Wootton JC (1994) Issues in searching molecular sequence databases. Nat Genet 6: 119-129. 7. Ferrer M, Martínez-Abarca F, Golyshin PN (2005) Mining genomes and ‘metagenomes’ for novel catalysts. Curr Opin Biotechnol 16: 588-593. 23. Wallace JC, Henikoff S (1992) PATMAT: a searching and extraction program for sequence, pattern and block queries and databases. Comput Appl Biosci 8. Krogh A, Brown M, Mian IS, Sjölander K, Haussler D (1994) Hidden Markov 8: 249-254. models in computational biology. Applications to protein modeling. J Mol Biol 235: 1501-1531. 24. Bailey TL, Gribskov M (1996) Proceedings of the Fourth International Conference on Intelligent Systems for Molecular Biology 15-24, Menlo Park, 9. Eddy SR, Mitchison G, Durbin R (1995) Maximum discrimination hidden California, AAAI Press. Markov models of sequence consensus. J Comput Biol 2: 9-23.

10. Wu C, Whitson G, McLarty J, Ermongkonchai A, Chang TC (1992) Protein 25. Lee SW, Won K, Lim HK, Kim JC, Choi GJ (2004) Screening for novel lipolytic classification artificial neural system. Protein Sci 1: 667-677. enzymes from uncultured soil microorganisms. Appl Microbiol Biotechnol 65: 720-726. 11. Wu CH, Berry M, Shivakumar S, McLarty J (1995) Neural networks for full- scale protein sequence classification: Sequence encoding with singular value 26. Yun J, Ryu S (2005) Screening for novel enzymes from metagenome and decomposition. Machine Learning 21: 177-193. SIGEX, as a way to improve it. Microb Cell Fact 4: 8.

J Data Mining Genomics Proteomics ISSN:2153-0602 JDMGP an open access journal Volume 3 • Issue 3 • 1000119